Defender project revisited

mecha

Well-known member
Joined
Apr 19, 2007
Messages
24,331
Reaction score
7,036
Location
MKE, Wisconsin
well, as a continuation of the same issue I had last year, I still have a Defender that plays the wrong sounds during gameplay. I have an assortment of new intel for this experiment of mine, so hold tight:

I will label parts as #1 (what came with the game) and #2 (what I bought extra)

sound board #2 has the [speech?] expansion slot, so I'm guessing that means it's a newer version?

this machine has a switching power supply installed with the old style RAM. the original power supply connector was not cut but instead was fitted with an adapter. I DO have an original Defender power supply that I bought along with a Bob Robert's repair kit, after reading that the Stargate one is the more desirable... I have not tested any voltages besides the +5, which I cannot remember where it's at because this has been an on-off again project for a year now.

1) I purchased a complete additional boardset. the CPU board boots up fine, the ROM board worked, however, had an assortment of roms installed that were in various shit conditions (missing pins/corrosion/etc.) so I wound up mixing and matching what I felt were the better looking roms between my 2 boards into one (potentially a mistake?)

2) sound board #1 would not work with all the #2 boards. in fact, the game would flash a green screen and have a white pattern screen, same condition as when the ROM board is unplugged from the CPU board. I used sound board #2 and the game booted, albeit playing incorrect sounds.

it should be worth noting I don't know what the Sound Test sound sounds like, if anyone has audio or video of this, please share it.

3) swapped back all known working roms to ROM board #1, sound board #1 now works. it plays entirely different sounds from sound board #2 that are also incorrect. swapped both socketed chips on the sound boards, same exact results.

4) I don't remember what condition changed when I did this, but I noticed that both sound boards had different dipswitches, and one board had both set to On and the other was set to Open (meaning off, yes?), I flipped the better working sound board #2 to whatever other option they were and I can't remember if there was a change or not.

5) to echo a previous thread of mine, I have swapped sound roms between the two boards with no change. refer to #1, did mixing my game roms cause this? they're both red label types dated 1981, with a few that have white labels dated 1980 (probably ones that didn't get changed between revisions?).

I don't have this game at my house or anything, it's at work, and I'm home now so I don't know what the dipswitch arrangements were. lol

questions I have:

-what do the dipswitches do, and what should they be set at?

-what do I need to make a new Test Panel for this? whoever had it last completely removed the test panel and cut all the wiring going to the coin door. I'm a little tired of having to hotwire it to get it out of the bookkeeping mode.

-which chip exactly is the PIA chip? I remember there being 2 chips that have to match or else it won't work right. I don't work with this game at all, so I don't know what all the exact parts are, I'm not a know-it-all, and I want anyone that replies to assume that I don't know anything about this game at all and will need some direction to make it work.

thank you.
 
Last edited:
So do you know if you are truly getting the wrong sound or just incomplete sounds?

I had a sound issue with my Robotron that was eventually fixed by repining the harness going between the ROM board and the sound board. I was getting some sound but it certainly wasn't the correct sound. I though it was the 'wrong' sounds as you mentioned when in fact it was just the absent of some sounds...thus making in sound like the wrong sound..make sense?

My suggestion would be...in this order

Go back to the original hardware...

Reflow all the .156 pins on the main PCB, Sound, and ROM boards (hell..do the iface boards also).

Verify 5.05ish at all 3 boards (grey wires)

Use the switching PS for testing as it probably more reliable. The OG PS can be rebuilt after you get your game up and running. If you using a switcher you won't have a ground to the coin door switches and they won't work. I'm guessing that's why they were hacked up

Verify continuity on the back of the sound board pin to the back of the ROM board pins. My issue was caused by a bad connection at the sound board..I repined the hardness between the ROM board and the SOUND board. You may be able to test this by wiggling the connector at the boards...while someone is playing

Once you know you have good conductivity between the boards (and still have the 'wrong' sounds) then start looking at the sound board.

Pick up a cap kit and a the IC (think it's a 6808..there's a million posts about it here) for the sound board from BR.

Did you ever have the correct sounds with the original hardware?
 
Last edited:
original hardware didn't boot at all, hence the replacement set. I've verified that it's not the ROM board ribbon cable, as I've gotten both boards to work with CPU board #2. I'm guessing #1 was stuck in reset, and I'm not a logic probe wizard to test that and stuff, I'm generally super busy all the time at work being the sole tech, I just happened to have some downtime yesterday to tinker with this again.

I just find it strange that the 2 sound boards have a different test sound, like mentioned, I don't know which one's the correct one lol. and I don't think I'll be able to try that in MAME unfortunately.

also, one sound board plays the power-on sound, the other doesn't. I wanna say board #2 plays it, board #1 doesn't. I'll definitely check these things when I come in, I might even get video on them too to post.

I don't remember wtf I did, but when I went back to the original sound board, I think I mentioned that it wouldn't boot the game at all, ROM board would flash 4 lights once, and do nothing. just for fun, I unplugged the ROM board from the CPU board to see if there was a change and it did the same thing, then I don't know what I did next, but I got it to boot, but like I said, no power-on sound, and the sounds were like almost missing entirely.

also, you might be onto something with the connectors/pins, I unfortunately have no .156 header pins left to change out, I might be stuck with trying a wire brush on them or something unless that's a bad idea. where do you buy .156 headers at? I have plenty of pins and plastic housings to make new connectors however, so I'll start testing continuity.
 
From memory, I think the test sound for Video Sound ROM 1 (Defender) is a repeated "extra ship" sound. The DIP switches should both be off/open for Defender. Switch 1 selects notes vs. sounds (I don't think it does much with this sound ROM), switch 2 is speech on/off and should only be on when a speech board is used (eg. Sinistar). The PIA chip is a 40 pin IC with the number 6821 on it. On the sound board, it's next to the CPU, which is a socketed 40 pin chip. The sound board, ROM board, and I/O board all use the same PIA. I had a sound board playing the wrong sounds a while back, the problem was corrosion on the header pins where the sound board connects to the ROM board. I replaced the header, and the sounds were fixed. The original Defender wiring harness used "insulation displacement" connectors - the wires are just pushed into the back of the connector, and the connector pin cuts through the insulation to make contact with the wire. These aren't the greatest connectors, and can work loose. If you have loose wires on the jumper between the sound and ROM boards, it could produce some wrong sounds - but if it was your only problem both of your sound boards would have the same wrong sounds. 0.156 headers and pins are available through Great Plains Electronics. You can ground the input pins on connector J3 (next to the test switch) yourself to test sounds. Pin 1 is a key (ie. cut/missing), pins 2-5 and 7 are all sound select pins. Different sounds are played when different sets of pins are grounded at the same time. BTW, having a speech board connector doesn't mean the sound board is newer - Gorgar was the first machine to use this sound board with a speech board, and it pre-dates Defender by a couple of years.
 
Last edited:
One more thing, rather than just hitting the test button to check the sound board, go into the Defender diags and use the sound test. Hitting the test button just grounds the NMI (non maskable interrupt) pin on the CPU, so it doesn't test the input stage of the board at all. The Defender sound diags ground the J3 pins the same as during gameplay, so the whole input path is tested. This will give you an idea of which/how many sounds may be missing.
 
I usually buy my pins and housings from mouser. However Bob Roberts has a nice Williams repin kit already made up..it's a kit to repin the entire game. It'll probably cost ya a buck or 2 more from him but it takes the guess work out of ordering the right stuff.
 
well, as a continuation of the same issue I had last year, I still have a Defender that plays the wrong sounds during gameplay. I have an assortment of new intel for this experiment of mine, so hold tight:

I will label parts as #1 (what came with the game) and #2 (what I bought extra)

sound board #2 has the [speech?] expansion slot, so I'm guessing that means it's a newer version?

The older version of the sound board was rectangular (about 9" x 4" IIRC). The Defender sound card had the speech card connector. Later ones had it removed (except for Sinistar).

this machine has a switching power supply installed with the old style RAM. the original power supply connector was not cut but instead was fitted with an adapter. I DO have an original Defender power supply that I bought along with a Bob Robert's repair kit, after reading that the Stargate one is the more desirable... I have not tested any voltages besides the +5, which I cannot remember where it's at because this has been an on-off again project for a year now.

1) I purchased a complete additional boardset. the CPU board boots up fine, the ROM board worked, however, had an assortment of roms installed that were in various shit conditions (missing pins/corrosion/etc.) so I wound up mixing and matching what I felt were the better looking roms between my 2 boards into one (potentially a mistake?)

There were two versions of the ROMs (red and green) and they are semi-interchangeable. I don't remember the details, but there were either 2 or 3 ROMs which changed).

2) sound board #1 would not work with all the #2 boards. in fact, the game would flash a green screen and have a white pattern screen, same condition as when the ROM board is unplugged from the CPU board. I used sound board #2 and the game booted, albeit playing incorrect sounds.

it should be worth noting I don't know what the Sound Test sound sounds like, if anyone has audio or video of this, please share it.

Make sure the sound ROM is Video Sound ROM #1.

Go into the diagnostics and let it run. It won't exercise all the sounds but it runs through the major ones.

3) swapped back all known working roms to ROM board #1, sound board #1 now works. it plays entirely different sounds from sound board #2 that are also incorrect. swapped both socketed chips on the sound boards, same exact results.

4) I don't remember what condition changed when I did this, but I noticed that both sound boards had different dipswitches, and one board had both set to On and the other was set to Open (meaning off, yes?), I flipped the better working sound board #2 to whatever other option they were and I can't remember if there was a change or not.

The switches should be off. That avoids routing the sounds to the non-existant sound card.

5) to echo a previous thread of mine, I have swapped sound roms between the two boards with no change. refer to #1, did mixing my game roms cause this? they're both red label types dated 1981, with a few that have white labels dated 1980 (probably ones that didn't get changed between revisions?).

I don't have this game at my house or anything, it's at work, and I'm home now so I don't know what the dipswitch arrangements were. lol

questions I have:

-what do the dipswitches do, and what should they be set at?

-what do I need to make a new Test Panel for this? whoever had it last completely removed the test panel and cut all the wiring going to the coin door. I'm a little tired of having to hotwire it to get it out of the bookkeeping mode.

You need 1 locking on/off switch and 2 momentary contact switches. The wiring diagram shows how to hook them up.

-which chip exactly is the PIA chip? I remember there being 2 chips that have to match or else it won't work right. I don't work with this game at all, so I don't know what all the exact parts are, I'm not a know-it-all, and I want anyone that replies to assume that I don't know anything about this game at all and will need some direction to make it work.

thank you.

The PIA chips are the 40 pins chips on the ROM card and the right hand one on the sound card (the one on the left is the 6808 CPU chip). Th PIA's are 6821 chips.

ken
 
update:

I redid the connector that goes from the ROM board to the sound board, and reflowed all the pins on both boards with no change. I guess it's a good sign that it's at least playing game sounds, but at the wrong times yes? maybe I'm "missing" sounds like people are saying. I don't know how to get into the Sound Test on this, it's incredibly difficult to navigate that bookkeeping menu while hotwiring blue and green wires lol

I haven't tested any voltages yet. if I remove the metal mounting plate with all the boards and take the switcher with, can I run this on like a U5000 monitor by chance? I was thinking maybe this would be a task for bringing to my house to work on since I don't really get around to doing anything at work.

should I just redo all of the connectors for the hell of it?
 
Last edited:
There were two versions of the ROMs (red and green) and they are semi-interchangeable. I don't remember the details, but there were either 2 or 3 ROMs which changed).
Don't forget the white ones! I have a working early style ROM board with the 10-chip white ROM set. It shares some ROM code with the green ROMs. One of the ROMs on that board was green, and thinking it was wrong I pulled it, dumped it and checked the checksum against the white ROM of the same number. It was the right chip, so I checked the green ROM checksum, and the two chips were identical.
 
yeah, FWIW, the MAME sets are labeled as the older Green/White Label and the newer Red Label.

I tested both sound boards again, the old one doesn't make any startup sound, and when you play in game it makes some very faint hums. upon looking at both sound boards side by side I noticed..... the crystal is MISSING on the old sound board. could that be causing that?

also if both my boards have Red/White label roms then that means they're interchangeable right? I mixed/matched them to get the best looking ones, as the ones that came with my new board were plastered in corrosion and solder and other shit. one had a leg broken off. at any rate, I don't think that would cause my sound problem right, that's completely independent of the Video Sound ROM #1? (both my sound boards have Sound ROM #1, I forgot to mention this)

EDIT: studied up on how to get to the Sound Test by playing with it in MAME. I think it's worth mentioning... the only sounds that I hear in the game are 29, 30 and 31.
 
Last edited:
It certainly would be nice if you had a known working defender to try that sound board in.

Do you have any other Williams games you can rob the 6808 from for testing? I would try switching the 6808s from board to board and see if anything changes. If your dead board is missing the crystal that's probably why you got nothing.. and you very well might have a good 6808 there..and a bad one in the partially working board.

Here's a site that had the board revs and decent pics to compare your sound board for missing parts

http://www.robotron-2084.co.uk/techwilliamshardwareid.html#defender

If it were me I'd take the partially working board...cap it..new 6808 (6802) and check\reburn the ROM. When I have an issue with something I prefer the shotgun method..nothing pisses me of more than a game going down when people are over. Especially something I've already 'fixed'..and you are dealing with abused 30 year old components.
 
yeah, FWIW, the MAME sets are labeled as the older Green/White Label and the newer Red Label.

I tested both sound boards again, the old one doesn't make any startup sound, and when you play in game it makes some very faint hums. upon looking at both sound boards side by side I noticed..... the crystal is MISSING on the old sound board. could that be causing that?

also if both my boards have Red/White label roms then that means they're interchangeable right? I mixed/matched them to get the best looking ones, as the ones that came with my new board were plastered in corrosion and solder and other shit. one had a leg broken off. at any rate, I don't think that would cause my sound problem right, that's completely independent of the Video Sound ROM #1? (both my sound boards have Sound ROM #1, I forgot to mention this)

EDIT: studied up on how to get to the Sound Test by playing with it in MAME. I think it's worth mentioning... the only sounds that I hear in the game are 29, 30 and 31.

White Label = Green label except for the prototype 10 ROM White label set. Red label != Green Label for all ROMs. I don't have the equivalence map handy so I can't tell you which ROMs match up. The Green ROMs, IIRC were upright only. The Red ROMs came out to support the cocktails.

If you are only getting some of the sounds, it probably means that the input circuitry is bad and is only reacting to some of the data lines. If you have a logic probe, you will need to check the interface chips to make sure the signals are getting through. Or it means the PIA on the ROM card is only driving some of the signals down the data line.

ken
 
well I've traded the socketed chips between sound boards. is the other big soldered chip supposed to be the CPU on the sound board?

only thing I don't like about the non-working old sound board of mine is that the test pattern sounds like a digital car revving repeatedly.... I don't think that's what it's supposed to sound like lol

maybe I'll try switching the crystal over to that and see what happens next. everything else on the game works decent, the joystick has some side to side play in it, but I'll worry about all that when I get the friggin thing to work first.

EDIT: oh, and I lack a logic probe, however, I have an oscilloscope that someone gave me a few years back.. lol not sure how it works or if it works, but I have one of those.
 
Last edited:
Without a crystal the board does *nothing*, as without that timing signal the cpu won't run. First thing I'd try is putting a crystal in that board (GPE may have them) and see if it works.

If you're using the same type of ROM (all reds or all greens) you can mix and match without a problem - just make sure you put the ROMs in the right place as the numbering is a bit screwy. http://arcadecontrols.com/BBBB/wmsrombd.html

The big 40 pin socketed chip is the CPU, and you may be having problems with the PIA which is the soldered (if it hasn't been replaced) 40 pin chip next to the cpu.

There's a troubleshooting flowchart that may help with this, you can get it here:
http://www.vidpin.com/tech_videorecap.asp It does have a couple of minor errors, but it's pretty useful.
 
Without a crystal the board does *nothing*, as without that timing signal the cpu won't run. First thing I'd try is putting a crystal in that board (GPE may have them) and see if it works.

If you're using the same type of ROM (all reds or all greens) you can mix and match without a problem - just make sure you put the ROMs in the right place as the numbering is a bit screwy. http://arcadecontrols.com/BBBB/wmsrombd.html

The big 40 pin socketed chip is the CPU, and you may be having problems with the PIA which is the soldered (if it hasn't been replaced) 40 pin chip next to the cpu.

There's a troubleshooting flowchart that may help with this, you can get it here:
http://www.vidpin.com/tech_videorecap.asp It does have a couple of minor errors, but it's pretty useful.

ah ok, I thought the PIA was the soldered one, but I kept hearing about the 6068 chip so I was confused.

I'll give it another whirl in a bit I guess.
 
ok, swapped the crystal from sound board #2 to sound board #1. I get NO sound from this sound board now. I reflowed all the headers and the fuse holders on it as well.

after that, I redid all the power connectors off the wiring harness. game still works fine, just zero sound with this board now. I felt around on chips and such, the big socketed chip gets warm, as well as the transistor with the double heatsink. I hear amplifier hiss from the speakers, so I can only assume that's working.

I guess tomorrow I'll trade the crystal back to board #2. I'll probably rig up a test panel with some extra buttons and a toggle switch as well so I can navigate to the sound test and get voltage readings.

I also never mentioned, it's running the original RAM (4116?) and I have a lithium battery mod from Bob Roberts I have yet to install. I figured the CPU board works as-is, I won't go tinkering with that just yet lol

any idea of what kind of voltages I should be looking for? I have to dig up a guide to show the RAM pinouts. what's the optimal way for testing the RAM, do you go for the ones farthest away from the connector?

also, my oscilloscope doesn't have probes with it, and I don't know where to get a logic probe from since it appears Radio Shack doesn't carry them anymore.
 
ok, swapped the crystal from sound board #2 to sound board #1. I get NO sound from this sound board now. I reflowed all the headers and the fuse holders on it as well.

after that, I redid all the power connectors off the wiring harness. game still works fine, just zero sound with this board now. I felt around on chips and such, the big socketed chip gets warm, as well as the transistor with the double heatsink. I hear amplifier hiss from the speakers, so I can only assume that's working.

I guess tomorrow I'll trade the crystal back to board #2. I'll probably rig up a test panel with some extra buttons and a toggle switch as well so I can navigate to the sound test and get voltage readings.

I also never mentioned, it's running the original RAM (4116?) and I have a lithium battery mod from Bob Roberts I have yet to install. I figured the CPU board works as-is, I won't go tinkering with that just yet lol

any idea of what kind of voltages I should be looking for? I have to dig up a guide to show the RAM pinouts. what's the optimal way for testing the RAM, do you go for the ones farthest away from the connector?

also, my oscilloscope doesn't have probes with it, and I don't know where to get a logic probe from since it appears Radio Shack doesn't carry them anymore.

You should be seeing +5V Regulated on pin 3, +12V unregulated on pin 2 and -12V unregulated on pin 8. Ground is pin 5.

The "transistor" is the voltage regulator that takes the +12V and generates the +5V used by the logic. The +5V on pin 3 is not used.

The Joust version of the schematic is better than the Defender version. The Joust lists the tespoint voltages as:
TP1 = +12V unregulated
TP2 = -12V unregulated
TP3 = Ground
TP4 = +5V regulated.

ken
 
yeah, I was wondering if those were the test points around there. they looked a little too convenient. I'm gonna move the crystal back to board 2, then I'll test voltages this time.

then the next order I guess is rigging up a wannabe test panel to do sound test.
 
ok, new updates. so upon inspecting things, I was getting ready to test voltages when I realized the switcher that was transplanted into my machine DOES NOT output -12V. whether that makes a huge difference or not, I don't know, I would imagine so if that's one of the voltage requirements right?

at any rate, I started rebuilding the Defender PS board I had, for all those that swear by cleaning PCBs with simple green and a hose, I commend you, because the board looks like new now.

as I'm doing this I got to wondering what the hell the 3rd connector on the board was for and wondering how you get AC to it, and uh... yeah, that was in fact lopped off in my cab. I have all the necessary molex parts to rebuild a new connector, except I don't know where all the wires hook up to the transformer, as the wires were all cut off.

the wiring diagram I have calls for some "green" wire too, which I saw coming off the line filter, I'm not sure if that has anything to do with it too but I found several lugs on the transformer that have soldered wires and I don't know what colors are what.

if there's anything else I need to know about for redoing this 4J1 connector, please let me know.

also, I wiggled the ROM board -> sound board harness while the game was running and got it to play entirely different sounds. I'm inclined to believe that's correct, I don't think there's anything wrong with that harness because all the continuities checked out (I'm guesstimating). I also followed one of those flowcharts where you unplug that connector entirely to see if the game still plays sounds and it didn't (which is a symptom of a bad ROM board I guess?)

EDIT: Translation: primary main objective = finding out where the 4J1 wires hook up on the transformer. need a diagram of what colored wires go where on the transformer to get power to this board. I also broke the -5 LED off like a dickhead, these are diodes though, does that mean I have to throw a new LED in there?
 
Last edited:
Back
Top Bottom